Statistical Approaches to the Study of Self-Organised Nanostructures
Publication Date
July 23, 2024
Creator
Abstract
Spin-casting solutions of colloidal nanoparticles onto solid substrates produces a diverse array of self-organised nanostructured patterns. Such patterns are regarded as far-from-equilibrium, relying on the complex dynamics of the solvent as the thin liquid film rapidly retracts from the surface. The observed ordering allows for varied statistical analysis upon atomic force microscopy images of the resulting patterns. An analytical toolbox utilising various statistical approaches is discussed and applied, supported by numerical simulations of the experiment, carried out using modified versions of the model developed by Rabani et al.
A systematic study of the effects of both gold nanoparticle concentration in solution and introduced surface heterogeneities was carried out, scanning resulting patterned morphologies with atomic force microscopy and evaluating with relevant statistical analysis. Co-deposited fullerenes and atomic force microscopy probe-induced oxidation aimed to disrupt and mediate the typical dewetting process of the thin solvent film on a highly local level, enhancing our understanding of the dynamic properties of the solvent. A pathway from synthesis to ambient surface characterisation of nanostructured photoactive dyads was additionally studied.
A series of synchrotron x-ray experiments on fullerene-encapsulating molecules addressed the extent to which encapsulated molecules are electrostatically screened and decoupled from their external environment when deposited on a metal substrate. In the observed absence of temperature dependence of molecule height above the substrate surface, experimental NIXSW data across the temperature range is used to assess the mobility of caged molecules, as well as reconcile results within a two-adsorption site framework.
A case for being critical at the point of data entry is demonstrated by the data-mining and re-utilisation of historic dewetting pattern images previously produced within the group, to form data sets to confidently evaluate new statistical analysis software. This is presented with documented development of transparent automated mass image processing software, highly sensitive to images of multi-layered nanostructures and the inherent noise in scanning probe microscopy, by means of a statistical approach utilising modern data science methods.
Item Type
ethesis
Thesis Type
PhD
Supervisors
Subjects (LC)
Associated Schools / Departments
School of Physics and Astronomy
eprints ID
77258
UoN Repository URI
Except where otherwise noted, this item's license is described as
File(s)![Thumbnail Image]()
Name
JEAH_01_2024_Thesis_MC.pdf
Type
Full-text
Description
Examined. PhD Thesis of JEA Hodgkinson with Minor Corrections
Size
104.2 MB
Format
Adobe PDF
Checksum (MD5)
48b2658086561005f87412d6a259d83f